Azure ML endpoint - getting predict_proba output

@Volodymyr Petryshyn While deploying the endpoint, did you update the scoring script with the one that has support for global parameters? I think by default it uses only Inputs. I think you can cross check this from your swagger file of the endpoint.

@Volodymyr Petryshyn Did you get a chance to check the above suggestion?

Thank you for your quick response,
when I registered and deployed model as endpoint,
the scoring file and deployment files are auto generated ,
when I download the model, there is scoring_file_v_2_0_0.py file that has this code that looks like support predic_proba output.
Do I need to modify it when I download it and upload it manually back to Azure ML and register it? Because auto generated endpoint code doesn't have predic_proba output.

def run(Inputs, GlobalParameters={"method": "predict"}):
    data = Inputs['data']
    if GlobalParameters.get("method", None) == "predict_proba":
        result = model.predict_proba(data)
    elif GlobalParameters.get("method", None) == "predict":
        result = model.predict(data)
    else:
        raise Exception(f"Invalid predict method argument received. GlobalParameters: {GlobalParameters}")
    if isinstance(result, pd.DataFrame):
        result = result.values
    return {'Results':result.tolist()}

@Volodymyr Petryshyn Yes, I think you need to do that. Following guidance from this SO thread, you need to upload this file in code + environment screen while deploying as an endpoint.

@Volodymyr Petryshyn I am curious to check if you are able to proceed further with the updated scoring script?